인터페이스는 현지화되어 있으며, Skill 소개와 설명 문서는 선택한 언어로 AI가 자동 번역합니다. 코드·명령어·경로·전문 용어는 원문을 유지하며 원작자 문서가 우선합니다.
Skills 둘러보기
데이터 및 연구 · SKILL.md

data-and-storage

data-and-storage

Manage Galaxy datatypes, metadata, dataset storage, object stores, file sources, data managers, tool data tables, and storage safety boundaries.

Independent GitHub snapshot · Automated classification · Original documentation governs

Skill 문서

GitHub

name: data-and-storage description: "Manage Galaxy datatypes, metadata, dataset storage, object stores, file sources, data managers, tool data tables, and storage safety boundaries." disable-model-invocation: true metadata: disco-role: operating license: NOASSERTION

Galaxy Data and Storage

Use this sub-skill when a task involves Galaxy datatypes, sniffers, datatype metadata, dataset storage, object stores, file sources, data managers, tool data tables, .loc entries, or storage cleanup risk assessment.

Start Here

  • For datatype implementation, registration, metadata, composite formats, and sniffer safety, read Data Formats.
  • For data manager XML/JSON contracts and tool-data table entries, read Data Managers.
  • For object-store, file-source, user template, credential, and storage-test planning, read Object Stores and Files.
  • For common failures and conservative triage, read Troubleshooting.
  • For a quick bounded-sniffer review, run python scripts/check_datatype_sniffer.py PATH_OR_STDIN.

Routing Boundaries

  • Stay here for datatype classes, datatypes_conf.xml registration, sniff, set_meta, set_peek, data-manager outputs, tool-data tables, object-store configuration concepts, file-source credentials, dataset cleanup safety, and data/storage tests.
  • Route broad Galaxy config placement, startup, galaxy.yml editing, and admin deployment planning to Configuration and Admin.
  • Route tool XML mechanics, wrapper tests, workflow YAML, and dependency declarations to Workflows and Tools.
  • Route upload/download API calls, history dataset automation, and API-client scripting to API Automation.

Safety Defaults

  • Never run cleanup, purge, migration, cloud object-store, or credentialed file-source actions without explicit user approval, target confirmation, backup/rollback context, and dry-run or read-only checks first.
  • Treat sniffers as production hot paths: they must inspect bounded prefixes or bounded header/sample helpers, never entire files.
  • Treat user file sources and object-store templates as secret-bearing configuration; do not print, persist, or invent credentials.
  • Prefer local/unit tests and template/model validation before remote integration tests; skip cloud or network tests unless credentials and disposable resources are explicitly supplied.
저장한 Skills